Is there a simple way (a function, a method...) of validating a character that a user types to see if it's compatible with Mac OS Roman? I've read a few dozen topics to find out why an iOS application crashes in reference to CGContextShowTextAtPoint. I guess an application can crash if it tries to draw on an image a string (i.e. ©) containing a character that is not included in the Mac OS Roman set. There are 256 characters in this set. I wonder if there's a better way other than matching the selected character one by one with those 256 characters?
Thank you
You might give https://developer.apple.com/library/mac/#documentation/graphicsimaging/conceptual/drawingwithquartz2d/dq_text/dq_text.html a closer read.
You can draw any encoding using CGContextShowGlyphsAtPoint instead of CContextShowTextAtPoint so you can tell it what the encoding is. If the user types it then you'll be getting the string as an NSString which is a Unicode string underneath. Probably the easiest is going to be to get the utf8 encoding of that user entered string via NSString's UTF8String method.
If you really want to stick with the very limited MacRoman for some reason, then use NSString's cStringUsingEncoding: passing in NSMacOSRomanStringEncoding to get a MacRoman string. Read the documentation on this in NSString though. Will return null if the user string can't be encoded in MacRoman losslessly. As it discusses you can use dataUsingEncoding:allowLossyConversion: and canBeConvertedToEncoding: to check. Read the cautions in the Discussion for cStringUsingEncoding: about about lifecycle of the returned strings though. getCString:maxLength:encoding: might end up being a better choice for you. All discussed in the class documentation for NSString.